From Copilot to Autonomous AI Agents | Power Automate Guide
summary
Shows how to turn Copilot-style assistants into agents that call Power Automate flows, connect to SharePoint, SQL and Word docs as custom knowledge, and trigger actions across 300+ connectors to do CRUD, approvals, emails and more. Demonstrates a weather agent that calls the Microsoft Weather API, stores user answers in variables, returns forecasts in chat and sends email via a Power Automate flow. Also builds apps with generative pages (uses GPT‑4.1 today; generative pages require a US environment) and emphasizes low‑code enterprise deployment in Microsoft Teams.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 1:02Copilot agents purpose
- 2:28AI agent misconceptions
- 3:52Copilot agents save time
- 5:03Define Copilot agent steps
- 6:27Copilot agents and databases
- 7:13Microsoft connectors benefits
- 8:16Power Automate repetitive tasks
- 9:38Custom knowledge sources
- 10:43Enable AI agent features
- 12:09Publish agents in Teams
- 13:32Agents boost productivity
- 15:36Customize agent responses
- 16:39Add custom topics
- 17:45Power Automate connectors
- 18:27Live demo Copilot agent
- 21:17Weather agent in Teams
- 22:24Trigger flows from agents
- 23:30Generative Pages intro
- 26:50Creating applications with Generative Pages
- 28:50Jira dashboard demo
- 30:16Closing and Q&A
related talks
46mHow work gets done in the Age of Al: Copilot, Apps and Agents
46mAI innovation in the Microsoft Power Platform
09mBuilding Apps with AI: Agents + Power Platform in Action
53mKeynote - Business Apps and Agents
25mAgent Flows: Build Agentic Automations You Can Trust
01h 03mAgents + SharePoint: The next generation of knowledge experiences
49mShaping next-gen development: the future of Copilot in Power Platform
25mAgentic User Interfaces in Copilot and Power Platform